1. Field of the Invention
This invention relates to networking of terminals based on a “peer-to-peer” architecture, and more particularly a method and system for accessing peer-to-peer networks.
2. Description of Prior Art
It is particularly but not exclusively applicable to downloading of files or sets of files provided by users of such networks.
Peer-to-peer network architectures enable two terminals connected to a transmission network such as the Internet network to communicate with each other directly without passing through a central server that redistributes the data. In this architecture, each terminal is configured so as to act as both server and client. In this manner, each terminal can share its information with an unlimited number of other terminals without needing to use a central server. Therefore, this architecture provides file exchange functions.
Some so-called centralized peer-to-peer networks are used by a server managing a database memorizing a user identifier for each network user, associated with a list of files provided to the other network users. Decentralized peer-to-peer networks also exist in which the database memorizing information about network users is distributed on some user terminals that agree to perform this database function.
At the present time, there are several peer-to-peer networks that are independent of each other such as OpenNap, Gnutella, Kazaa and eDonkey. Each of these networks is used in the Internet network by installing a special access software to the peer-to-peer network on each terminal. The result is that a user cannot access such a network from any terminal. In particular, such an access would be difficult to envisage with a mobile telephone with a limited memory capacity and an operating system that is incompatible with the access software.
Furthermore, due to their enormous success, peer-to-peer networks tend to be saturated, such that when a user of such a network makes a request to download a given file, his request is frequently inserted in a long queue. The user must then often remain connected to the network for several hours before his request can finally be processed. Therefore, this type of downloading cannot be done from a mobile telephone, which firstly does not have a permanent connection to a network (not invoiced by the connection time) and secondly has insufficient memory capacities to receive the downloaded file.